A critical authorization vulnerability exists in the Joomla extension Gridbox versions prior to 2.20.4.2. The EditorController::getBlogPosts() task lacks the usual user edit level check, allowing unauthorized users to invoke this function without proper permission verification.
A Cross-Site Scripting (XSS) vulnerability exists in the processing of SVG images submitted to the /api/v2/ai/sessions API when using the data:image/svg+xml;base64 format. Embedded script tags in the SVG are executed when rendered in the assistant window, enabling JavaScript execution and content manipulation.
The twofactor_gauthenticator plugin for Roundcube has a vulnerability that allows brute-forcing of the second-factor authentication code due to lack of rate limiting. An attacker who has obtained a user's password can repeatedly submit arbitrary 6-digit codes to bypass the second-factor authentication and gain unauthorized access. This issue has been fixed in a specific commit.
An unauthenticated SQL injection vulnerability exists in the Joomla extension Gridbox versions prior to 2.20.4.2. The vulnerability occurs on store product list pages where the 'sort-by' URL parameter is improperly handled, allowing injection into the ORDER BY clause of a SQL query without proper validation.
A critical stack-based buffer overflow vulnerability exists in JP1/Automatic Job Management System 3 - Manager versions from 10-00 up to but not including 13-50-31. This vulnerability allows an attacker to execute arbitrary code with no privileges required and no user interaction needed.
A stack-based buffer overflow vulnerability exists in JP1/Automatic Job Management System 3 - Manager versions from 10-00 up to but not including 13-50-31. This vulnerability has a critical severity with a CVSS score of 9.4, indicating it can be exploited remotely without privileges or user interaction to cause high confidentiality and integrity impact, and low availability impact.
A critical stack-based buffer overflow vulnerability exists in JP1/Automatic Job Management System 3 - Manager and Agent components, affecting versions from 11-00 up to but not including 13-50-31. This vulnerability allows for remote code execution without privileges or user interaction.
A stack-based buffer overflow vulnerability exists in JP1/Automatic Job Management System 3 affecting both Manager and Agent components from version 11-00 up to but not including 13-50-31. The vulnerability has a high severity with a CVSS score of 9.4, indicating it can be exploited remotely without privileges or user interaction, leading to high confidentiality and integrity impact and low availability impact.
A Cross-Site Request Forgery (CSRF) vulnerability exists in the category management functionality of the Phoca Download Joomla extension versions 3.1.0 through 6.1.5. This vulnerability could allow an attacker to perform unauthorized actions on behalf of an authenticated user. The issue is classified as medium severity with a CVSS score of 5.3.
The twofactor_gauthenticator plugin for Roundcube has a vulnerability that allows bypassing two-factor authentication (2FA) by spoofing client address headers. An attacker who has obtained a user's password can spoof whitelisted IP addresses via the Client-IP and X-Forwarded-For headers, bypassing the second authentication factor. This issue has been fixed in commit 4640889.
A heap-based buffer overflow vulnerability exists in JP1/Automatic Job Management System 3 affecting both Manager and Agent components from version 11-00 up to but not including 13-50-31. The vulnerability has a critical severity with a CVSS score of 9.4, indicating it can be exploited remotely without privileges or user interaction, potentially leading to high confidentiality and integrity impact and low availability impact.
Docker Desktop versions from 4.39.0 up to but not including 4.97.0 could log third-party API keys and OAuth access tokens in cleartext within local application logs and diagnostic bundles due to insufficient inter-process communication payload redaction. This exposure allows an attacker with read access to these files to obtain sensitive credentials. Versions 4.94.0 and 4.95.0 partially addressed the issue by fixing known logging paths, but residual logs containing credentials may persist after upgrading. The issue is fully resolved in Docker Desktop 4.97.0, which also cleans up existing host logs and crash dumps containing sensitive data.
CVE-2026-107590 is an out-of-bounds write vulnerability in the Qt QTextCodec iconv-based encoding within the Qt 5 Compatibility APIs module. Under certain conditions, encoding text with QTextCodec may write outside an internal buffer, potentially causing application crashes and affecting data integrity. The vulnerability affects Qt versions from 4.5.0 up to and including 6.8.8, and from 6.9.0 up to and including 6.11.2. The CVSS 4.0 base score is 1.9, indicating low severity. No public exploits or patches are currently documented.
NVIDIA PhysX contains an out-of-bounds write vulnerability that could allow an attacker to execute code, tamper with data, cause denial of service, or disclose information. This vulnerability affects all platforms running PhysX. The flaw is identified as CWE-787 and has a high severity score of 7.8. No specific affected versions or patches are currently detailed.
CVE-2026-65145 is a critical vulnerability in the NVIDIA NeMo Agent Toolkit involving missing authentication for a critical function. This flaw allows an attacker to perform unauthorized actions without authentication, potentially leading to code execution, privilege escalation, data tampering, denial of service, and information disclosure.
CVE-2026-65151 is a high-severity SQL injection vulnerability in the NVIDIA NeMo Agent Toolkit. Exploitation could allow an attacker to tamper with data, cause denial of service, or disclose sensitive information. No affected versions or patch information are provided.
ICMacOS (Icarus macOS) is a macOS malware-as-a-service stealer marketed since September 2026 by a Russian-speaking actor. It targets macOS Sierra through current versions on both Intel and Apple Silicon architectures. Initially a credential stealer, it evolved into a persistent remote access trojan (RAT) with WebSocket-based command-and-control. Infection uses social engineering with fake system dialogs to harvest administrator passwords and deploy the payload. Version 1 steals browser credentials, cryptocurrency wallets, developer secrets, messaging data, and password managers. Version 2 adds RAT features with persistence and interactive shell access, expanding data theft to Steam accounts, Outlook data, and more password managers.
Two unrelated threat operations exploited domain shadowing techniques to insert illegal gambling content under legitimate educational and government domains in Bangladesh, India, and Mexico. Attackers gained unauthorized DNS administration access, creating CNAME records and delegating subzones to third-party nameservers, leveraging institutional domain reputations for search engine optimization. The campaigns affected over 850 entry points across eight domains, primarily hosting gambling pages on Cloudflare Pages. These operations served illegal gambling content targeting Indonesian and Indian markets, exploiting institutional trust to boost search rankings and traffic while leaving primary websites intact and potentially unnoticed by domain owners.
NVIDIA TensorRT-LLM version 1.3.0rc24 contains a server-side request forgery (SSRF) vulnerability. This flaw could allow an attacker to induce the server to make unauthorized requests, potentially leading to information disclosure. The vulnerability has a high severity rating with a CVSS score of 7.1. There is no indication of known exploits in the wild or available patches at this time.
CVE-2026-24269 is a server-side request forgery (SSRF) vulnerability in NVIDIA TensorRT-LLM version 1.3.0rc24. Exploitation could allow an attacker to cause unauthorized requests from the server, potentially leading to information disclosure and data tampering. The vulnerability has a high severity score of 7.1 CVSS v3.1. No public exploits or patches are currently documented.
